
The Geometry of Generalized Binary Search
This paper investigates the problem of determining a binaryvalued funct...
read it

Simple sufficient condition for inadmissibility of Moran's singlesplit test
Suppose that a statistician observes two independent variates X_1 and X_...
read it

On Universality and Training in Binary Hypothesis Testing
The classical binary hypothesis testing problem is revisited. We notice ...
read it

Interpretable hypothesis tests
Although hypothesis tests play a prominent role in Science, their interp...
read it

Optimal ROC Curves from Score Variable Threshold Tests
The Receiver Operating Characteristic (ROC) is a wellestablished repres...
read it

The Structure of Optimal Private Tests for Simple Hypotheses
Hypothesis testing plays a central role in statistical inference, and is...
read it

Stay Together: A System for Single and Splitantecedent Anaphora Resolution
The stateoftheart on basic, singleantecedent anaphora has greatly im...
read it
Generalized Binary Search For SplitNeighborly Problems
In sequential hypothesis testing, Generalized Binary Search (GBS) greedily chooses the test with the highest information gain at each step. It is known that GBS obtains the gold standard query cost of O( n) for problems satisfying the kneighborly condition, which requires any two tests to be connected by a sequence of tests where neighboring tests disagree on at most k hypotheses. In this paper, we introduce a weaker condition, splitneighborly, which requires that for the set of hypotheses two neighbors disagree on, any subset is splittable by some test. For four problems that are not kneighborly for any constant k, we prove that they are splitneighborly, which allows us to obtain the optimal O( n) worstcase query cost.
READ FULL TEXT
Comments
There are no comments yet.